Safactory: A Scalable Agentic Infrastructure for Training Trustworthy Autonomous Intelligence
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915990843949056 |
|---|---|
| author | Chen, Xinquan Yin, Zhenyun He, Shan Huang, Bin Lei, Shanzhe Shi, Pengcheng Cai, Kun Chen, Bei Liu, Bangwei Kang, Zeyu Huang, Chao Zhang, Yang Li, Wenjie Ge, Ruijun Wang, Yajie Fang, Tianshun Xu, Tianyang Cong, Yiwen Jin, Meng Li, Gaolei Wu, Xuansheng Liu, Linhan He, Zijing Li, An Teng, Yan Tan, Xin Liu, Dongrui Shao, Jing Lu, ChaoChao He, Ji Li, Jie Song, Chunfeng Xu, Jinya Song, Fan Wang, Shujie Qian, Jianmin Hou, Jie Wang, Xuhong Wang, Yingchun Wang, Hui Hu, Xia |
| author_facet | Chen, Xinquan Yin, Zhenyun He, Shan Huang, Bin Lei, Shanzhe Shi, Pengcheng Cai, Kun Chen, Bei Liu, Bangwei Kang, Zeyu Huang, Chao Zhang, Yang Li, Wenjie Ge, Ruijun Wang, Yajie Fang, Tianshun Xu, Tianyang Cong, Yiwen Jin, Meng Li, Gaolei Wu, Xuansheng Liu, Linhan He, Zijing Li, An Teng, Yan Tan, Xin Liu, Dongrui Shao, Jing Lu, ChaoChao He, Ji Li, Jie Song, Chunfeng Xu, Jinya Song, Fan Wang, Shujie Qian, Jianmin Hou, Jie Wang, Xuhong Wang, Yingchun Wang, Hui Hu, Xia |
| contents | As large models evolve from conversational assistants into autonomous agents, challenges increasingly arise from long-horizon decision making, tool use, and real environment interaction. Existing agenticinfrastructure remain fragmented across evaluation, data management, and agent evolution, making it difficult to discover risks systematically and improve models in a continuous closed loop. In this report, we present \textbf{Safactory}, a scalable agent factory for trustworthy autonomous intelligence. Safactory integrates three tightly coupled platforms: a \textbf{Parallel Simulation Platform} for trajectory generation, a \textbf{Trustworthy Data Platform} for trajectory storage and experience extraction, and an \textbf{Autonomous Evolution Platform} for asynchronous reinforcement learning and on-policy distillation. As far as we know, Safactory is the first framework to propose a unified evolutionary pipeline for next-generation trustworthy autonomous intelligence. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_06230 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Safactory: A Scalable Agentic Infrastructure for Training Trustworthy Autonomous Intelligence Chen, Xinquan Yin, Zhenyun He, Shan Huang, Bin Lei, Shanzhe Shi, Pengcheng Cai, Kun Chen, Bei Liu, Bangwei Kang, Zeyu Huang, Chao Zhang, Yang Li, Wenjie Ge, Ruijun Wang, Yajie Fang, Tianshun Xu, Tianyang Cong, Yiwen Jin, Meng Li, Gaolei Wu, Xuansheng Liu, Linhan He, Zijing Li, An Teng, Yan Tan, Xin Liu, Dongrui Shao, Jing Lu, ChaoChao He, Ji Li, Jie Song, Chunfeng Xu, Jinya Song, Fan Wang, Shujie Qian, Jianmin Hou, Jie Wang, Xuhong Wang, Yingchun Wang, Hui Hu, Xia Artificial Intelligence Distributed, Parallel, and Cluster Computing As large models evolve from conversational assistants into autonomous agents, challenges increasingly arise from long-horizon decision making, tool use, and real environment interaction. Existing agenticinfrastructure remain fragmented across evaluation, data management, and agent evolution, making it difficult to discover risks systematically and improve models in a continuous closed loop. In this report, we present \textbf{Safactory}, a scalable agent factory for trustworthy autonomous intelligence. Safactory integrates three tightly coupled platforms: a \textbf{Parallel Simulation Platform} for trajectory generation, a \textbf{Trustworthy Data Platform} for trajectory storage and experience extraction, and an \textbf{Autonomous Evolution Platform} for asynchronous reinforcement learning and on-policy distillation. As far as we know, Safactory is the first framework to propose a unified evolutionary pipeline for next-generation trustworthy autonomous intelligence. |
| title | Safactory: A Scalable Agentic Infrastructure for Training Trustworthy Autonomous Intelligence |
| topic | Artificial Intelligence Distributed, Parallel, and Cluster Computing |
| url | https://arxiv.org/abs/2605.06230 |